Hostname: page-component-7b7567c8c-tnzcf Total loading time: 0 Render date: 2026-11-13T10:06:24.633Z Has data issue: false hasContentIssue false
  • English
  • Français

RELIGION, TAXES, AND SEX DISCRIMINATION:

Where Do Liberal Principles Lead?

Published online by Cambridge University Press:  08 September 2005

Andrew Altman
Andrew Altman
Affiliation:
Georgia State University
Rights & Permissions [Opens in a new window]

Extract

Modern liberalism developed out of the strife of post-Reformation religious warfare. Among liberalism's central ideas were those of the individual's right of religious liberty and the separate jurisdictions of secular and religious authority. In societies that accepted these ideas and put them into institutional practice, levels of systemic religious violence were dramatically diminished. Moreover, the liberal principles that helped to build and sustain civil peace could make a strong claim to providing a just framework for addressing religious differences. Yet important normative questions have remained about the policies a liberal state should adopt toward religion.

Information

Type
ARTICLES
Information
Legal Theory , Volume 11 , Issue 2 , June 2005, pp. 125 - 142
Copyright
© 2005 Cambridge University Press

I. INTRODUCTION

Modern liberalism developed out of the strife of post-Reformation religious warfare. Among liberalism's central ideas were those of the individual's right of religious liberty and the separate jurisdictions of secular and religious authority. In societies that accepted these ideas and put them into institutional practice, levels of systemic religious violence were dramatically diminished. Moreover, the liberal principles that helped to build and sustain civil peace could make a strong claim to providing a just framework for addressing religious differences. Yet important normative questions have remained about the policies a liberal state should adopt toward religion.

Many of the questions still under debate revolve around whether and when religious institutions and individuals ought to be exempt from legal requirements that unintentionally burden the exercise of their religion. Of particular importance are questions concerning the extent to which religious institutions ought to be exempt from antidiscrimination laws. Such laws have been reasonably seen as serving important social interests and protecting basic individual rights. But many liberal thinkers who strongly support antidiscrimination laws nonetheless balk at the idea that liberal principles require the laws to be applied willy-nilly to religious institutions.

It is widely agreed that these institutions ought to be exempt from legal liability for the ways in which they fill jobs that have religious functions. Such an exemption is recognized under the laws of existing liberal states and is known as the “ministerial exemption” in U.S. legal doctrine.1

U.S. courts have consistently held that the ministerial exemption is required by the Free Exercise and Establishment clauses of the First Amendment. See McClure v. Salvation Army, 460 F.2d 553 (5th Cir. 1972); and EEOC v. Catholic University 83 F.3d 455 (D.C. Cir. 1996). The U.S. Supreme Court has not ruled directly on the issue. In Australia, human rights legislation explicitly grants a ministerial exemption. See, e.g., the Anti-Discrimination Act of 1977, No 48, New South Wales, as updated to July 1999, s. 56 (a). In Canada, the general approach is to give religious institutions the right to invoke a defense of “Bona Fide Occupational Requirement” (BFOR) when their employment practices run contrary to legal antidiscrimination norms. See Alvin J. Esau, “Islands of Exclusivity”: Religious Organizations and Employment Discrimination, 33 U. B.C. L. REV. 33 (2000) 719. However, in Nova Scotia, legislation explicitly grants a ministerial exemption; see the Human Rights Act. R.S.N.S. 1989, as amended 1991, c. 214, s. 6 (c) iii. The term “ministerial exemption” clearly reflects the dominant Protestant culture of the United States. It should be understood as a synecdoche standing for a range of spiritual positions, although some courts have also construed it to cover the position of church choir direction. See Miller v. Bay View United Method Church 141 F. Supp. 2d 1174 (ED Wisconsin, 2001). The term “church” should similarly be construed as a synecdoche for institutions of religious worship in general. It is important to distinguish such institutions from schools that have a religious affiliation. See infra, note 6.

The dominant view among legal and political theorists is that liberal principles require such an exemption.2

See, e.g., Martha Nussbaum, A Plea for Difficulty, in IS MULTICULTURALISM BAD FOR WOMEN? 114 (Susan Moller Okin, ed., 1999); BRIAN BARRY, CULTURE AND EQUALITY 174–176 (2001); Kent Greenawalt, Freedom of Association and Religious Association, in FREEDOM OF ASSOCIATION 118 (A. Gutmann, ed., 1998); and WILLIAM GALSTON, LIBERAL PLURALISM 111 (2002).

A number of prominent scholars have gone beyond this view to argue that religious institutions that discriminate on the basis of sex in filling ministerial offices ought not to be, ipso facto, disqualified from tax-exempt status.3

Under current U.S. law, churches generally qualify for an exemption to federal income taxes under I.R.C. §501 (c) (3). Also, state and local laws exempt religious institutions as nonprofit organizations from property taxes.

We can describe the position of these scholars as the “full-exemption position.”4

To clarify, “full exemption” does not mean that a tax exemption for religious institutions that discriminate in ministerial hiring is required, but only that such discrimination must not automatically disqualify such an institution from an exemption.

In this article, I examine and assess the arguments made for the full-exemption position from liberal principles. My analysis will show that the arguments are not persuasive. Instead, the best liberal arguments support a position that endorses the ministerial exemption but also holds that it is permissible for the state to disqualify religious institutions that discriminate on grounds of sex in ministerial hiring from receiving tax-exempt status.5

It is generally agreed that discrimination by religious institutions on the basis of an individual's religion is a legitimate exception to employment discrimination laws. See, e.g., 42 U.S.C.A. §2000e-1 (2001) (provision in Title VII of the Civil Rights Act of 1964, providing that the antidiscrimination prohibitions of the law do not apply to religious institutions “with respect to the employment of individuals of a particular religion”). The Supreme Court has ruled that, insofar as the Title VII provision covers nonministerial positions, there is no violation of the Establishment Clause. Corporation of the Presiding Bishop of the Church of Jesus Christ of Latter-Day Saints v. Amos 483 US 327 (1987). See also the Manitoba case, Schroen v. Steinbach Bible College (1999) 35 C.H.R.R. D/1 (upholding under the BFOR rationale the firing by a Mennonite school of an accounting clerk who was not a Mennonite). The U.S. Supreme Court's reasoning is criticized in Nancy Rosenblum, Amos: Religious Autonomy and the Moral Uses of Pluralism, in OBLIGATIONS OF CITIZENSHIP AND DEMANDS OF FAITH 165–195 (N. Rosenblum, ed., 2000).

Sections II to VI explain and criticize the arguments for the full-exemption position. Section VII turns to the alternative view that the state ought to have a ministerial exemption but that it is also required to disqualify from tax-exempt status religious institutions that discriminate on the basis of sex in their ministerial hiring. The arguments for this view are found wanting. The section also examines and finds inadequate the position that there must be no ministerial exemption for the practice of male-only ordination and that institutions following the practice must be disqualified from receiving tax-exempt status. In Section VIII, I argue that liberal principles require the ministerial exemption and permit (but do not require) the denial of tax exemptions to religious institutions that discriminate on grounds of sex in their ministerial hiring.6

The focus of this paper is on institutions of religious worship and their hiring practices for spiritual positions. Issues concerning discrimination at religious schools in policies that pertain to students raise separate issues. See Bob Jones University v. U.S. 461 U.S. 574, 604 (1983) (holding that a university with a religious affiliation could be denied tax-exempt status because of its racially discriminatory dating policy, even when the policy had a religious rationale). The court distinguished educational institutions with a religious affiliation from institutions of worship (at 604). Amy Gutmann defends the dominant view among liberal thinkers, endorsing the Bob Jones holding in Gutmann, ed., Freedom of Association: An Introductory Essay, in FREEDOM OF ASSOCIATION 7 (1998). Michael McConnell defends a minority view, criticizing the holding in McConnell, Free Exercise Revisionism and the Smith Decision, U. CHI. L. REV. 57 (1990) 1146.

II. THE RELEVANCY PRINCIPLE

Brian Barry argues for the full-exemption position by claiming that the same liberal principle that requires a ministerial exemption also bars the state from withholding tax-exempt status from institutions because they engage in sex discrimination when filling ministerial offices. The principle in question holds that a person is treated unjustly if she is denied an employment opportunity on the basis of some characteristic of hers that is not relevant to her qualifications for the job. Let's call this the “relevancy principle.”7

BARRY, supra note 2, at 174. The term “relevancy principle” is mine, not Barry's. The relevancy principle is itself a corollary of a more basic liberal commitment to principles of impartiality and rationality, both of which make the formal demand that decisions be grounded on relevant considerations.

A corollary of the principle is embodied in existing law as the “BFOQ” exception to antidiscrimination norms. The exception permits an employer to discriminate on the basis of sex if it can be shown that sex is “a bona fide occupational qualification [BFOQ] reasonably necessary to the operation of an employer's business.”8

U.S. v. City of Buffalo, W.D.N.Y.1978, 457 F.Supp. 612, affirmed as modified 633 F.2d 643. It should be noted that under existing U.S. case law, the BFOQ exception does not apply to discrimination based on race. Ray v. University of Arkansas, E.D.Ark.1994, 868 F.Supp. 1104.

Barry argues that the relevancy principle and its BFOQ corollary demand that the state refrain from applying its antidiscrimination norms to ministerial offices and from withholding tax-exempt status because of discriminatory ministerial hiring policies.

When it comes to ministerial employment, Barry contends, sex can be a bona fide occupational qualification. If the practices and doctrines of a religious institution embody the idea that only men can effectively administer the institution's sacred rites and rituals, then sex will be a relevant qualification. He asks: “[H]ow can [a person's sex] be irrelevant if the people concerned believe it to be of the essence? If you believe that the sacraments have efficacy only if administered by a man, you can scarcely regard the sex of the person administering them as irrelevant.”9

BARRY, supra note 2, at 174. The case of the ministerial exemption is one of the few in which Barry endorses what he calls the “rule-and-exception” approach to formulating fair legal standards. His view is that such an approach generally fails on grounds of fairness and that the application of uniform legal rules to all persons is crucial for treating all persons as equals. See BARRY, supra note 2, at 40–50.

Barry argues that the relevancy principle also bars the state from withholding tax-exempt status from religious institutions because they discriminate based on sex in their ministerial employment. Were the state to adopt a policy of denying tax exemptions for that reason, then it would be offering a benefit to religious institutions on the condition that they fill ministerial offices with persons who are not qualified by the standards of those institutions. But an offer by the state with such a condition attached is fundamentally illegitimate, in Barry's view. The offer would foster the violation of another corollary of the relevancy principle: it is an injustice to qualified applicants when an institution hires unqualified ones. Accordingly, Barry concludes: “If churches are to get favourable treatment [i.e., tax-exempt status] at all, their doing so should not be contingent on their abandoning their position on the necessary qualifications for holding religious office.”10

BARRY, supra note 2, at 168.

III. RELEVANCY AND THE MINISTERIAL EXEMPTION

Let us begin the assessment of Barry's argument with his defense of the ministerial exemption. It will turn out that certain problems with the defense spillover into the second part of his argument, where he addresses the issue of tax exemptions.

In his argument for the ministerial exemption, Barry elides an important distinction. It is one thing for a group of people to believe that an action, such as administering a sacrament, will not have the intended consequences if the person performing it is a woman. It is another thing for such a belief to be true. The relevancy principle concerns what is true about the effects of the action and not simply what people believe to be true about the effects.

At the same time, liberal principles prohibit the state from making or acting on judgments about the truth or falsity of religious doctrines, such as those concerning the efficacy of sacraments. This prohibition is part of the minimal content of the principle of church-state separation.11

See U.S. v. Ballard 322 U.S. 78 (1944).

Indeed, the prohibition is what bars the state from justifying a ministerial exemption on grounds of the theological assertion that excluding women from ministerial offices is not unjustly discriminatory because God regards the roles that women fill within the church as equal in importance to the roles of men.

Yet it is precisely this prohibition against the state passing judgment on religious claims that creates a problem with the first stage of Barry's argument. The state must refrain from judgments about the truth of the belief that sacraments or other religious rites have the intended consequences only if performed by men. Accordingly, if the issue turns on actual effects—whether, for example, rites of baptism wash away original sin or conducting mass effects transubstantiation, the state must suspend judgment on the question of whether the relevancy principle prohibits the exclusion of women from the priesthood.

One might think that, given the state's lack of authority to render judgment about the actual efficacy of sacramental rites, the liberal position must be to defer to a church's view that sex is relevant to its spiritual offices. But such thinking rests on a non sequitur. The correct conclusion is, rather, that the relevancy principle cannot and should not be applied to claims about the supernatural effects of actions. Accordingly, liberal principles of church-state separation render the relevancy principle useless for arguing that sex is a BFOQ for the efficacious administration of the sacraments.

However, there is another way to understand Barry's BFOQ argument for the ministerial exemption.12

I am indebted to Eric Brown for suggesting this way of reading the argument.

On this alternative, it is not necessary for the state to affirm or assume the truth of any theological claim about the efficacy of religious rites. Instead of asking the state to accept its doctrines as true, a church simply needs to ask the state to accept that certain doctrines are those of the church itself, apart from the question of their truth, and that belief in those doctrines is an essential qualification for its ministerial offices. So a church seeking to defend its male-only ministry can reasonably assert that anyone who does not endorse the belief in male-only ordination lacks an essential qualification. But in that case a woman who sought ordination would be stuck in a pragmatic self-contradiction if she also were to claim that she endorsed male-only ordination.13

Barry writes of a hypothetical case of a woman who sues the Roman Catholic Church to become a priest: “She claims…to be a Roman Catholic, which entails acceptance of Papal authority. Yet at the same time she is asking a court to overrule the Pope's decision about women priests.” BARRY, supra note 2, at 176.

The woman's act of seeking ordination would belie her claim to believe in the church's doctrine regarding ordination, and so we could justifiably conclude that she lacked an essential qualification for the job.

The problem with this argument is that it begs the crucial question of whether it is legitimate for a liberal state to ban, in the first place, a type of ministerial job in which sex is admittedly a BFOQ. This problem derives from the fact that the relevancy principle and its BFOQ corollary do not directly address questions about whether a given kind of job, with its particular qualifications, should be permitted or prohibited by the liberal state.

Liberal states justifiably ban a range of types of jobs, at least indirectly, by banning activities that are essential to the job. The pertinent considerations for the state do not directly involve the principle that the employment opportunities of a person should depend on her job qualifications. Rather, the considerations involve such factors as the harm or risk of harm to others, the need for solving coordination problems, and other variables that are admissible as grounds for the liberty-restricting laws of a liberal state.

If the state were illegitimately to ban some category of jobs—say the state was acting on invalid paternalist reasons in its ban—then individuals who were qualified for the job would have a good argument that they were being unjustly denied employment opportunities. The type of job in question should be legally permitted, and if it were permitted, the relevancy principle would require that the qualified applicants get the jobs, or at least that such applicants have job prospects that reflect how well-qualified they are. But that argument is clearly parasitic upon a prior argument that the ban was illegitimate in the first place.

A ban on sex discrimination in ministerial employment is essentially a ban on a certain kind of job category: male-only clergy. The state would be declaring that a certain essential qualification makes a certain kind of job illegitimate so long as the job is constituted in a way that makes that qualification essential. To escape the ban, the job-kind must be redefined so that sex is no longer an essential qualification. The liberal state may in fact lack the authority for banning the male-only clergy and demanding that churches change the way in which they construct their ministerial job-kinds. But the BFOQ argument does not show that to be the case. Some independent argument is needed to show why such a ban would be illegitimate.

Applying the relevancy principle straightaway to a sex discrimination claim is justifiable only if we can assume that the job and its conventional qualifications have been constituted in a way that does not reflect or reinforce sex-based injustice.14

Cf. IRIS YOUNG, JUSTICE AND THE POLITICS OF DIFFERENCE 212 (1991). Young rejects the “myth of merit” (at 200ff), but she relies on a version of the relevancy principle in formulating her own standards for fair employment opportunity. She writes, “Criteria should not exclude any social groups from consideration for positions, either explicitly or implicitly.” Clearly, Young has an implicit understanding of what counts as a relevant social group, according to which groups defined in terms of sex would be included but not a group defined in terms of, say, literacy for the position of lawyer or librarian.

When the rules that constitute the position and its qualifications reflect and perpetuate background sex-based injustice, then a simple appeal to the principle will not do.

IV. TAX EXEMPTIONS AND LIBERTY

Barry's use of the relevancy principle to conclude that tax-exempt status ought not be withheld from institutions on the ground of their discriminatory ministerial hiring faces the same kind of problem that undercuts his reasoning on the ministerial exemption. The application of the principle is parasitic on some more basic argument concerning the considerations for which a liberal state may grant or withhold favorable tax treatment. Just as there are certain job-kinds that a liberal state could legitimately ban precisely on account of the essential qualifications for the job-kind as conventionally constituted, there are certain job-kinds such that a liberal state could legitimately withhold favorable tax treatment from institutions whose workplace incorporates such positions. The relevancy principle and BFOQ rule do not address the question of which job-kinds may form the basis for the withholding of tax-exempt status.

At the same time, it should be conceded that withholding tax-exempt status from churches because of the way they define their ministerial offices is an infringement on the liberty of the members of those churches. The tax policy imposes on the institutions a certain condition: you can get the exemption but only if your qualifications for the ministry are altered in a way dictated by us, the state. Although this policy may not involve an infringement on liberty as serious as would be constituted by a flat-out ban on the male-only priesthood, it does have the effect and intent of imposing a cost on the exercise of religious liberty by the members of churches that do not permit the ordination of women.

It might be argued that there really is no punitive intent in withholding tax-exempt status, but even if that were so, the consideration would not be decisive. It is doubtful that a punitive intent is a necessary condition for an act to count as an infringement on liberty. Some of the most egregious violations of liberty involve not the intent to punish but rather the paternalistic intent to help. For example, consider the traditional restrictions on the liberty of women, such as their exclusion from the legal profession, which were adopted for essentially paternalistic reasons.15

See Bradwell v. Illinois 83 U.S. 130 (1872).

In a real-world scenario, the policy of withholding tax exemptions would be generally construed as conveying the authoritative condemnation of the practice of a male-only ministry. This is especially so because the status quo is that religious institutions, including discriminatory ones, enjoy such status: removing tax exemptions from discriminatory churches but leaving them in place for the rest would be bound to be interpreted as official condemnation of the churches that lose their exemptions. Even if official condemnation were not the intent, the fact that the policy was generally understood that way would make it a liberty-restricting policy.

Moreover, it is difficult to conceive of a real-world scenario in which the intention to communicate the authoritative repudiation of the male-only ministry would not be among the intentions of those supporting the policy. Indeed, it would likely be a primary intent. Of course, the supporters might also be intending to “keep the hands of the state clean” by withdrawing a kind of state subsidy from practices they regard as complicit in sex-based injustice.16

Cf. Henry Cline, The Priority of Democratic Autonomy over Discriminatory Religion, J. PHIL. RES. 25 (2000) 396, 398.

But it would be odd for someone to intend to keep the state's hands clean from certain practices without also intending to send an authoritative message that the practices were “dirty” ones that merited condemnation. The message would be an essential part of the account given by the supporters of why the withdrawal of the subsidy was considered the right thing to do. And when a practice is authoritatively condemned by official action, the liberty to engage in it has been restricted in a significant way.

Because the male-only priesthood is a job-kind that is likely the product of traditional conceptions of sex that unjustly disadvantage women, there is reason to think that the liberal state ought to withdraw tax-exempt status from discriminatory religious institutions, notwithstanding the liberty-restricting character of the policy. But, again, the liberties that are restricted hold a special place in liberal thought, and it is possible that there is a reasonable liberal argument based on the status of those liberties against the withholding of tax exemptions from discriminatory churches.

Accordingly, instead of arguing from the relevancy principle, it seems that those who favor the full-exemption position ought to argue that religious and expressive liberty trump sex equality, at least in the context of ministerial employment. Nonetheless, the question of which values hold the trump card can be a vexed one, and it might be thought that a more straightforward route to the full-exemption position could rest on the second key principle that guided the liberal settlement of early modern religious conflict, that is, the principle of church-state separation.

V. SEPARATION OF CHURCH AND STATE

The exact contours of church-state separation are subject to considerable controversy. Nonetheless, liberal thought presumes that some line between the two jurisdictions can be drawn that is nonarbitrary and reasonably determinate.17

Stanley Fish argues that liberal efforts to draw the line between church and state “can have no justification apart from the act of power performed by those who determine the boundaries.” His argument rests on an uncritical acceptance of the premise that the only real reasons are prudential ones, so that liberal reasoning “from principle” is always and only a matter of “political” calculation. See Fish, Mission Impossible: Settling the Just Boundaries of Church and State, COLUM. L. REV. 97 (1997) 2261, 2271.

One view holds that the best understanding of church-state separation locates the authority to determine criteria for ministerial offices wholly beyond the proper jurisdiction of the state and exclusively within that of the church. This view is endorsed by Barry as well as by Michael McConnell, among other prominent advocates of the full-exemption position, and by courts in the United States.18

See BARRY, supra note 2, at 175–176; and Michael McConnell, The Problem of Singling out Religion, DEPAUL L. REV. 50 (2000) 19–20. In Gellington v. Christian Methodist Episcopal Church, Inc. 203 F.3d 1299, 1304 (11th Cir 2000), the court concluded, “The Establishment Clause…mandates that churches retain exclusive control over strictly ecclesiastical matters.”

Accordingly, Barry describes the matter of who is to be eligible for ministerial offices as “purely internal” to a church. Any dispute among church members about criteria for eligibility is one that the state may not step in to settle, because such a step would involve the illegitimate “substitution of secular for religious authority.”19

BARRY, supra note 2, at 176.

McConnell endorses much the same view: “churches and synagogues have an absolute right to select their spiritual leaders without being subject to second-guessing on behalf of the state, even when the conduct would otherwise violate anti-discrimination laws.”20

McConnell, supra note 18, at 19–20.

Once the ministerial exemption is justified on the basis of such an understanding of church-state separation, it becomes difficult to avoid the conclusion that churches should not be denied tax-exempt status because they engage in sex discrimination in their ministerial hiring. Given churches' “absolute right to select their spiritual leaders,” it seems clear that a state would be exceeding its proper jurisdiction if it were to condition tax-exempt status on meeting antidiscrimination norms. The transgression would derive not from the mere denial of tax-exempt status but from the reason for the denial. Because the reason concerns qualifications for the ministry, the state would be determining its tax policies toward religious institutions on the basis of their essentially religious practices and would thereby be exceeding its secular jurisdiction.

Against the church-state argument, it might be countered that the state would be acting within its jurisdiction if it were simply denying tax exemptions: such exemptions are a privilege to which neither churches nor any other institutions have a right. Tax-exempt status is legitimately given to an institution or withheld from it based on the state's judgment about how that status would advance important social purposes. If the state denies the status to churches that discriminate in their ministerial hiring, then it is acting on a judgment that it is within the jurisdiction of the state to make, namely, that no sufficiently important social purpose would be served by providing the churches with such a privilege.

However, this counterargument fails if it does not also reject the Barry-McConnell premise that the matter of ministerial qualifications is a “purely internal” one for the church over which it has an “absolute right.” Suppose that a state were to make the judgment that no sufficiently important purpose is served by extending tax-exempt status to churches that engage in sex discrimination in their hiring. In arriving at the judgment, it would be necessary for the state to take account of and evaluate by its own secular criteria the ministerial employment practices of churches. After all, churches that do not discriminate would, ex hypothesi, receive tax-exempt status, so the only ground on which the status could be denied to discriminatory churches would be the judgment that the bad of the discrimination outweighs the good that the churches otherwise do. But in reaching such a conclusion, the state would not be treating ministerial qualifications as a purely internal matter for the church. Rather, the state would be claiming a legitimate, albeit limited, interest of its own in the nature of those qualifications. So if ministerial hiring really is a strictly internal church matter, then it is difficult to see how to avoid the Barry-McConnell position on tax exemptions.

VI. ASSESSING THE CHURCH-STATE ARGUMENT

The premise that ministerial qualifications are a purely internal church matter is, however, a dubious one. The logic of the Barry-McConnell argument can be turned against the premise itself. For it seems that there are clear instances, albeit rare ones, in which a state is justified in denying tax-exempt status to churches on account of their ministerial qualifications. By the argument's logic, then, the Barry-McConnell premise about the churches' jurisdiction must be wrong.

Suppose that a church made it a necessary condition for holding a position in its ministry that an individual participate in a ritual human sacrifice. State prosecution of participants in such a ritual would, of course, not only be permissible but obligatory. That fact by itself does not show that the ministerial qualifications are within the state's jurisdiction: the ritual sacrifices would amount to murder and should be prosecuted on that basis, regardless of their relation to the church's ministerial qualifications. Thus the state would not be basing its prosecutorial policy on the qualifications as such. Rather, what shows that the ministerial qualifications are within the state's jurisdiction is the fact that it would be permissible and even obligatory for the state to deny any such church tax-exempt status precisely because the church required people to commit murder in order to become ministers.

It might be countered that the hypothetical case of ministerial murder demonstrates that nothing short of the most compelling state interest—the protection of human life—can justify the state in treating a church on the basis of its qualifications for the ministry. But consider a church committed to no-growth environmental principles and suppose that it required individuals to destroy property in new suburban developments in order to be eligible for its ministry. It would seem clear that the state ought to deny any such church tax-exempt status on the basis of its ministerial qualifications. The protection of private property is an important state interest but it does not amount to the level of importance of the protection of life. Moreover, it makes no difference if the rules of the hypothetical church specify that the property damage should be worth no more than, say, $500. Such a church should still be denied tax-exempt status.

It seems, then, that there is no straightforward route from the principle of church-state separation to the full-exemption view of how the liberal state ought to treat religious institutions that exclude women from ministerial offices. More generally, the principle does not appear to provide an argument independent of considerations of the importance that a liberal state ought to attach to religious liberty.21

Cf. Ira Lupu, Free Exercise Exemption and Religious Institutions: The Case of Employment Discrimination, B.U. L. REV. 67 (1987) 414.

VII. ALTERNATIVE VIEWS

In light of the problems with the arguments for the full-exemption position, it seems reasonable to take an alternative view and argue that the ministerial exemption is required to protect religious liberty but that churches that practice male-only ordination ought to be denied tax-exempt status. Pursuing such an argument, Ian Shapiro contends that government may justifiably “attach some costs” to religious institutions “in which some offices are reserved for men, persons of a particular race, or any other group defined in a morally arbitrary way.” Sanctions such as the withholding of tax-exempt status “can, over time, be expected to encourage civil institutions to evolve in comparatively democratic directions.”22

IAN SHAPIRO, DEMOCRACY'S PLACE 244 (1996).

For Shapiro, a more democratic direction is one that diminishes the extent to which the institutions of civil and political society disadvantage persons based on morally arbitrary distinctions such as sex and race. Accordingly, his argument for the policy of denying tax-exempt status to discriminatory churches rests on the idea that such a policy will help move society in that way. However, Shapiro fails to explain why the ministerial exemption should not also be denied in order to encourage society to evolve in a more democratic direction. Removing the exemption would also impose a cost on discriminatory religious institutions and arguably encourage society to move in a democratic direction, and some account needs to be given for why that cost may not be imposed but the cost of withholding tax-exempt status may be imposed.

Some people may suggest that the key difference lies in the fact that withholding tax-exempt status merely involves denying a benefit, not imposing a cost. However, Shapiro does not appear to take this suggestion, and, in any event, it will not stand scrutiny. Even conceding that a mere denial of benefit is involved in withholding tax-exempt status, the crucial question is whether the members of discriminatory churches possess, as a matter of their religious liberty, the right not to be denied that benefit on account of the ministerial hiring practices of their institution. If they have that right, it matters not whether the situation is described as denying a benefit or imposing a cost. Shapiro's argument assumes that there is no such right, but the assumption begs a crucial question against the proponents of the full-exemption position.23

Shapiro also fails to make explicit the conditions that must be met in order for infringements on religious liberty to be justified. Shapiro writes that “a substantial degree of religious autonomy of citizens from the state is an important value in a modern democracy,” and he then proceeds to argue that religious freedom can be trumped in certain situations, e.g., when parents exercise it to thwart the basic interest that children have in getting an education adequate for them to function as democratic citizens. However, cases of parental rights with respect to children do not help to shed much light on the issues of the ministerial and tax exemptions. See SHAPIRO,supra note 22, at 138, 152 (from chapter coauthored with Richard Arneson). Jeffrey Spinner-Halev appears to take a position very similar to Shapiro's with respect the exemption issues, but he also neglects to spell out the conditions he regards as necessary (or sufficient) for the justifiable infringement on religious liberty. See JEFFREY SPINNER-HALEV, SURVIVING DIVERSITY: RELIGION AND DEMOCRATIC CITIZENSHIP} 176, 182, 187 (2000).

Henry Cline regards a tax-exemption as a form of state subsidy and holds that when the state provides a subsidy to a religious institution that practices male-only ordination, it is putting “a stamp of government approval on…discriminatory attitudes.”24

Cline, supra note 16, at 398.

Cline contends that the basic political principle of free and equal citizenship prohibits “subsidizing discriminatory religion.”25

Cline, supra note 16, at 397.

On this analysis, individuals have no right that the state subsidize their religious institutions when those institutions engage in sex discrimination.

However, Cline's argument fails for several reasons. First, one cannot coherently maintain that a state that grants tax exemptions to institutions that discriminate based on sex thereby puts its stamp of approval on sex discrimination. The liberal state also gives exemptions to institutions that reject sex discrimination. By Cline's logic, we would also have to conclude that the state is putting its stamp of approval on nondiscriminatory ministerial hiring.

More important, the ministerial exemption is an indirect subsidy for discriminatory institutions: it permits them to escape civil penalties for otherwise illegal practices. If the direct financial subsidy of tax-exempt status is prohibited by the principle of free and equal citizenship, then it would appear that the indirect subsidy of a ministerial exemption would also violate the principle. Yet Cline does not question the validity of the ministerial exemption. He might claim that religious individuals have a right that their institutions be given a ministerial exemption but not tax-exempt status. However, such a claim begs the question against advocates of the full-exemption view. As with Shapiro's argument, we are left with no non–question-begging argument for an alternative to the full-exemption position.

Jane Rutherford takes the principle of equal citizenship to forbid the ministerial exemption and to bar as well tax-exempt status for institutions that practice male-only ordination.26

Jane Rutherford, Equality as the Primary Constitutional Value: The Case for Applying Employment Discrimination Laws to Religion, CORNELL L. REV. 81 (1996) 1049.

Her view is thus the mirror image of the full-exemption position. However, Rutherford simply asserts that equality is incompatible with either form of exemption. In doing so, she fails to take account of arguments made by William Galston, Martha Nussbaum, and others that the idea of equal citizenship in a liberal state does not mean that each and every institution in civil society must embody the state's public norms of equality. Public institutions may certainly not bar women from certain employment positions. However, it would seem that individuals—including women—have a right voluntarily to become and remain members of religious institutions that practice male-only ordination.

Accordingly, Nussbaum argues that in such liberal democracies as the United States, “religious organizations are voluntary” and that “it seems illiberal to hold that practices internal to the conduct of the religious body itself—the choice of priests, the regulations concerning articles of clothing—must always be brought into line with a secular liberal understanding of the ultimate good.”27

Nussbaum, supra note 2, at 113–114. Galston holds a similar view on this point: “While we may regret the exclusion of women from the Catholic priesthood and from the rabbinate of Orthodox Judaism, I take it that we would agree that otherwise binding antidiscrimination laws should not be invoked to end these practices…. [R]eligious associations (and perhaps others as well) enjoy considerable authority within their own sphere to determine their own affairs.” See WILLIIAM GALSTON, LIBERAL PLURALISM 111 (2002).

At the same time, she holds that religions should be required to conform to laws guaranteeing “equal liberty and opportunity to all adult citizens in fundamental areas of life,” including “the opportunity to seek employment outside the home.”28

GALSTON,supra note 26.

In opposition to Nussbaum's view, it might be argued that an indirect but important effect of having a ministerial exemption is to make it more difficult for women to be treated as equal citizens in those fundamental areas of life where liberal principles demand equality. And it is reasonable to think that male-only ordination policies work in some degree against the public norm of equality.

However, if there is a right to religious liberty, then any restriction on its exercise must, at a minimum, bear a burden of persuasion. In the case of the ministerial exemption, that burden would be reasonably understood as involving the requirement to produce clear and convincing evidence that the denial of the exemption would significantly improve women's status as citizens in those areas of life having fundamental importance for citizens. Yet there is no such evidence. It is plausible to hypothesize that denying the exemption would lead some institutions of worship to change their male-only rule of ordination, which would in turn help to equalize the status of women in those areas of importance to which Nussbaum refers. Yet no one really has a good idea of how likely it is that such a hypothesis is true or the degree to which sex equality would be helped. In light of the uncertainties, we lack adequate grounds to justify a rather significant incursion on the right of religious liberty. Thus it is difficult to escape the conclusion that if there is a right of religious liberty at all, then it protects institutions of worship from the requirement to conform their ministerial hiring to legal rules against sex discrimination.29

In recent years, there has been growing skepticism about whether religious freedom should be treated as a basic liberty on a par with freedom of speech and the press. Some scholars have suggested that treating religious liberty as basic undermines the idea of equality under the law. Christopher L. Eisgruber and Lawrence G. Sager, Religious Liberty and the Moral Structure of Constitutional Rights, 6 LEGAL THEORY 267 (2000). It seems to me that this skepticism rests on an overly rigid understanding of the principle of equality under the law and that some reasonable accommodation is possible between the claims of conscience (religious or secular), on the one side, and the idea that all citizens should be equal under the law, on the other. The claims of conscience need not be denied even if they cannot be understood as strictly parallel to freedom of speech and the press. Much of the current debate on religious liberty stems from the case Employment Division, Oregon v. Smith 494 U.S. 872 (1990) (holding that an infringement on the free exercise of religion must meet the most exacting legal standard only if the infringement also restricts some other, basic constitutional liberty or if the infringement is imposed on a case-by-case determination).

The question remains whether the right of religious liberty protects religious institutions from having tax-exempt status denied to them on account of their male-only ministerial hiring practices. In order to get a handle on the matter, it is necessary to look more closely at the nature of such a status and the proper role of government in granting it.

VIII. DISCRETION AND DISCRIMINATION

Tax-exempt status is best understood as a discretionary benefit offered by government to advance important societal goals. To say that the benefit is discretionary is to say that it is permitted but not required. Government is not prohibited from offering the benefit for the purpose of advancing the goal but it is not required to offer the benefit, even conceding that offering it does advance the goal. Perhaps some policy or other must be adopted by government in light of the importance of the goal.30

One could say that government has an “imperfect obligation” to advance the social goal in question, in one sense of that term. See J.B. SCHNEEWIND, THE INVENTION OF AUTONOMY 78, 132–133 (1998).

However, tax-exempt status is not the kind of benefit that government is required to provide: the choice to offer the benefit or not is within the government's discretionary authority.

Nonetheless, even a discretionary choice is required to respect moral principles; otherwise the choice amounts to an abuse of discretion. Thus potential recipients of discretionary benefits may not be excluded from receiving them on arbitrary grounds or for reasons of wrongful prejudice. If tax-exempt status is given to churches in one part of the country, government must not deny it to other churches because they are located in a different part of the country. If tax-exempt status is given to Christian houses of worship, government must not deny it to Jewish ones for reasons of anti-Semitism.

Religious institutions that practice male-only ordination might argue that it would constitute an abuse of discretion if government were to deny them tax-exempt status while giving the status to institutions that ordain women and men. On this argument, such a governmental policy would be wrongfully discriminatory, akin to providing tax-exempt status to Christian but not Jewish institutions of worship.

However, such an argument is not convincing. As Cline notes, the policy of denying of tax-exempt status to discriminatory religious institutions would be a symbolic affirmation of the principle that sex should make no difference when it comes to the rights and responsibilities of citizenship.31

Cline, supra note 16, at 396.

And as Shapiro argues, the policy might also be part of an effort to provide incentives for institutional change that would result in a fuller realization of the principle that sex should make no difference when it comes to citizenship.32

Shapiro, supra note 22, at 244.

This principle of sex equality is central to the liberal state, and there is nothing wrongful in affirming the principle or promoting its fuller realization by denying a discretionary benefit to those institutions whose policies work against it. Policies based on anti-Semitic rationales or motives are a wholly different matter. Those policies cannot be reasonably understood as affirmations of the liberal principle of equal citizenship.

It may be true that a policy of denying tax-exempt status to institutions practicing male-only ordination would place a burden on those institutions, one that would not be borne by other religious institutions. However, whether the policy amounts to wrongful discrimination cannot be judged simply on the basis of the effects of a policy when what is at stake is the distribution of some discretionary benefit. All such policies will have differential effects, disadvantaging some people relative to other people, and in judging whether wrongful discrimination is involved, it is crucial to take into account the reasons behind the policy.33

For this reason, I reject the “disparate-impact” conception of discrimination in favor of a “disparate treatment” approach, for cases where a discretionary benefit (or burden) is involved. See below in this section for a discussion of cases in which a basic right is involved. I should add that courts in the United States have interpreted the disparate treatment approach too narrowly by requiring (for cases invoking the Equal Protection Clause of the Fourteenth Amendment) that specifiable individuals have an intention to harm the victims before a claim of discrimination can be sustained. See Washington V. Davis 426 U.S. 229 (1976). Indifference on the part of the public generally to the interests of the members of certain social groups can lead to discrimination just as much as intentional hostility by identifiable individuals to the interests of the group members.

Discriminatory policies involving discretionary benefits are ones whose motivating reasons involve an arbitrary or irrational discounting of the interests of persons who are detrimentally affected.

It is important to distinguish cases in which a discretionary benefit is involved from those in which some basic right is at stake. When policies violate the basic rights of some but not all citizens, then it is tempting to describe the situation as one of wrongful discrimination. However, the fundamental wrong in such a case is not the discrimination, that is, it is not the differential treatment of citizens, but rather the violation of the right. Thus citizens have a basic right to vote, and a policy of denying it to racial minorities is wrong not because the members of the racial majority have their right to vote respected but rather because the policy transgresses the rights of minorities. And the wrong in question does not depend in any way on the motives or reasons behind the denial.

In contrast, when discretionary benefits such as tax exemptions are involved, no one has an unconditional right to the benefit, that is, a right that is independent of whether other people are getting the benefit. In such cases, the motives or reasons that explain why some people are getting the benefit and others are not receiving it become central to the question of whether it is a case of wrongful discrimination. Thus governmental policies that withheld tax exemptions from certain institutions or persons based on anti-Semitic beliefs would count as wrongfully discriminatory. On the other hand, a policy that withheld such a benefit in order to affirm symbolically or more fully realize the liberal state's principle of equal citizenship would not count as an instance of wrongful discrimination. No arbitrary or irrational discounting of anyone's interests motivates the latter policy, in contrast to the anti-Semitic case.

It is possible to imagine a situation in which denying tax-exempt status to religious institutions that practice male-only ordination does amount to wrongful discrimination, and so my claim is not that such a denial never wrongfully discriminates. Suppose, for example, that only the Roman Catholic Church did not ordain women and that the rest of society harbored substantial anti-Catholic prejudice. Also suppose that this prejudice was the reason for denying tax-exempt status to institutions that did not permit female ordination. In such a case, the relative disadvantage imposed on some religious institutions would amount to wrongful discrimination.

One might claim, however, that to deny tax-exempt status to institutions that practice male-only ordination is ipso facto to be motivated by prejudice against the people who belong to them. There seems to be no difference between the hypothetical anti-Catholic case and any other case of denying tax-exempt status to institutions that do not permit female ordination. However, the claim is not correct, and there is a difference.

Consider the anti-Catholic hypothetical again. How do we know that it is a case of wrongful prejudice rather than a permissible affirmation of liberal equality? The key is to look for a pattern of policies that systematically disadvantage Catholics. The belief that the interests of Catholics count for less than those of others is not likely to be reflected in only one policy. Rather, there is likely to be a series of policies that, taken together, cannot be reasonably explained on any premise other than that anti-Catholic prejudice is operating to influence the law. The pattern proves the prejudice.

Yet there is no pattern of policy-imposed disadvantage when it comes to religious institutions in contemporary liberal societies that practice male-only ordination. It is logically possible that there be prejudice without the pattern or a pattern without the prejudice in a particular case. But in the absence of a pattern of disadvantage, the presumption should be that the members of institutions that ordain only males are not the victims of a form of bigotry akin to anti-Semitic or anti-Catholic prejudice. And it is difficult to see what empirical factors exist under current conditions that would rebut such a presumption. In the absence of such rebuttal evidence, a policy of denying tax-exempt status could only be reasonably understood as an affirmation and promotion of equality citizenship regardless of sex. As such, the policy would not amount to wrongful discrimination.

But one might still ask how the interests of the members of institutions that practice male-only ordination are to be properly taken into account. The answer is that the ministerial exemption gives those interests their due weight. The members have a right that the ministerial exemption be part of antidiscrimination law. They have no such right when it comes to tax-exemption because it concerns a discretionary benefit and government has legitimate reasons for denying them the benefit.

On the other hand, Cline goes too far in arguing that government must deny the benefit.34

Cline, supra note 16, at 96–98.

The government of a liberal state has many options for ratifying and promoting the principle of sex equality. Granting the right to vote, regardless of sex, is not optional but mandatory, because such a right is a basic claim to which each citizen is entitled. However, denying tax-exempt status to sex-discriminatory institutions of worship is an eligible option but it is not required.

Consider the analogy with “bias-crime laws,” that is, criminal laws that provide extra punishment for crimes committed as a result of prejudiced or discriminatory motives. Laws in this category typically pick out certain crimes and say that those who commit them because of the victim's race or sex or ethnicity shall be eligible for a greater punishment than someone committing the same crime from a different motive, such as pecuniary gain.

Frederick Lawrence has defended these laws, arguing that “a society dedicated to equality must treat bias crimes differently from other crimes and must enhance the punishment of these crimes.”35

FREDERICK LAWRENCE, PUNISHING HATE: BIAS CRIMES UNDER AMERICAN LAW} 100 (1999).

He points out that bias-crime laws constitute an authoritative “societal condemnation of racism, religious intolerance and other forms of bigotry.”36

Id. at 167.

Such bigotry certainly runs contrary to the principle of equal citizenship. However, it does not follow that a society dedicated to equality must punish bias crimes more severely than the underlying crime. Assuming that such treatment for bias crimes is a legitimate option for a liberal society, the society still does not have any obligation to use the criminal law in that way.37

Kahan argues that many aspects of criminal law in liberal society are analogous to imposing extra punishment for bigoted motives and that there is nothing illegitimate about using the criminal law in such a way. In contrast, Hurd argues that using the criminal law to punish bigoted motives violates basic liberal principles because it amounts to punishing people for having a bad character. I think that Kahan's analysis is closer to the mark, at least under conditions of a liberal state in which there are still systemic violations of the principle of equal citizenship and the reasons for the bias-crime laws concern the promotion or affirmation of equality. Bias-crime laws are a modest measure that may well advance equality and, under such conditions, do not violate the basic rights of criminals. See Dan Kahan, Two Liberal Fallacies in the Hate Crimes Debate, 20 LAW & PHIL. 175–193 (2001); and Heidi Hurd, Why Liberals Should Hate “Hate Crime Legislation,” 20 LAW & PHIL. 215–232 (2001).

Lawrence makes a mistake parallel to the one made by Cline: both believe that a particular affirmation of equality is mandatory, when in fact the affirmation is part of an array of eligible options for advancing equality. Society is obligated to take some reasonable steps to realize more fully the principle of equal citizenship but it is not obligated to enact bias crime laws or withdraw tax-exempt status from religious institutions that ordain only males. Each of those policies remains within the discretionary power of the liberal state.

IX. CONCLUSION

The formative ideas of liberalism revolved around matters of religious liberty and tolerance. Centuries of philosophical reflection and institutional innovation have settled many important questions about the stance a liberal state ought to take toward religion. The outstanding questions may not be as urgent as the life-and-death matters that first drove the development of liberalism. However, the questions that remain do pose important social and philosophical problems.

Religious institutions that practice male-only ordination confront the liberal state with such problems. The practice amounts to unjust sex discrimination by the public norms of liberalism. However, the practice is also an instance of the exercise of religious liberty. What stance, then, should the liberal state take toward these religious institutions and their practices? In this paper, I have argued that liberal principles lead to the conclusion that the state must provide a ministerial exemption to such institutions and that it is at liberty to extend or deny tax-exempt status to the institutions. Such a conclusion accommodates the demands of religious liberty while freeing the liberal state to affirm and promote sex equality.